• Anderson, S. P., R. A. Weller, and R. B. Lukas, 1996: Surface buoyancy forcing and the mixed layer of the western Pacific warm pool: Observations and one-dimensional model results. J. Climate,9, 3056–3085.

  • Bony, S., K.-M. Lau, and Y. C. Sud, 1997: Sea surface temperature and large-scale circulation influences on tropical greenhouse effect and cloud radiative forcing. J. Climate,10, 2055–2076.

  • Brown, R. G., and C. Zhang, 1997: Variability of midtropospheric moisture and its effect on cloud-top height distribution during TOGA COARE. J. Atmos. Sci.,54, 2760–2774.

  • Clement, A. C., R. Seager, M. A. Cane, and S. E. Zebiak, 1996: An ocean dynamical thermostat. J. Climate,9, 2190–2196.

  • Esbensen, S., 1978: Bulk thermodynamic effects and properties of small tropical cumuli. J. Atmos. Sci.,35, 826–831.

  • Fu, R., A. D. DelGenio, W. B. Rossow, and W. T. Liu, 1992: Cirrus-cloud thermostat for tropical sea-surface temperatures tested using satellite data. Nature,358, 394–397.

  • Grabowski, W. W., 2000: Cloud microphysics and the tropical climate:Cloud-resolving model perspective. J. Climate,13, 2306–2322.

  • Graham, N. E., and T. P. Barnett, 1987: Sea surface temperature, surface wind divergence, and convection over the tropical oceans. Science,238, 657–659.

  • Johnson, R. H., 1978: Cumulus transports in a tropical wave composite for phase-III of GATE. J. Atmos. Sci.,35, 484–494.

  • Lau, K.-M., and C. H. Sui, 1997: Mechanisms of short-term sea surface temperature regulation: Observations during TOGA COARE. J. Climate,10, 465–472.

  • ——, ——, M. D. Chou, and W. K. Tao, 1994: An inquiry into the cirrus-cloud thermostat effect for tropical sea-surface temperature. Geophys. Res. Lett.,21, 1157–1160.

  • ——, H.-T. Wu, and S. Bony, 1997: The role of large-scale atmospheric circulation in the relationship between tropical convection and sea surface temperature. J. Climate,10, 381–392.

  • Liao, X., and D. Rind, 1997: Local upper tropospheric/lower stratospheric clear-sky water vapor and tropospheric deep convection. J. Geophys. Res.,102, 19 543–19 557.

  • Lucas, C., and E. J. Zipser, 2000: Environmental variability during TOGA COARE. J. Atmos. Sci.,57, 2333–2350.

  • Newell, R. E., 1979: Climate and the ocean. Amer. Sci.,67, 405–416.

  • Nicholls, S., and M. A. Lemone, 1980: The fair weather boundary layer in GATE: The relationship of subcloud fluxes and structure to the distribution and enhancement of cumulus clouds. J. Atmos. Sci.,37, 2051–2067.

  • Nilsson, J., and K. A. Emanuel, 1999: Equilibrium atmospheres of a two-column radiative-convective model. Quart. J. Roy. Meteor. Soc.,125, 2239–2264.

  • Numaguti, A., R. Oki, K. Nakamura, K. T. N. Misawa, T. Asai, and Y. M. Kodama, 1995: 4–5-day-period variation and low-level dry air observed in the equatorial Western Pacific during the TOGA-COARE IOP. J. Meteor. Soc. Japan,73, 267–290.

  • Ramanathan, V., and W. Collins, 1991: Thermodynamic regulation of ocean warming by cirrus clouds deduced from observations of the 1987 El Nino. Nature,351, 27–32.

  • Randall, D. A., and G. J. Huffman, 1980: A stochastic model of cumulus clumping. J. Atmos. Sci.,37, 2068–2078.

  • Raymond, D. J., 2000: The Hadley circulation as a radiative-convective instability. J. Atmos. Sci.,57, 1286–1297.

  • Sherwood, S. C., 1999: Convective precursors and predictability in the tropical western Pacific. Mon. Wea. Rev.,127, 2977–2991.

  • Simpson, J., 1980: Downdrafts as linkages in dynamic cumulus seeding effects. J. Appl. Meteor.,19, 477–487.

  • Sud, Y. C., G. K. Walker, and K.-M. Lau, 1999: Mechanisms regulating sea-surface temperatures and deep convection in the tropics. Geophys. Res. Lett.,26, 1019–1022.

  • Sun, D.-Z., and Z. Y. Liu, 1996: Dynamic ocean–atmosphere coupling: A thermostat for the tropics. Science,272, 1148–1150.

  • Tompkins, A. M., 2000: The impact of dimensionality on long-term cloud-resolving model simulations. Mon. Wea. Rev.,128, 1521–1535.

  • ——, 2001: Organization of tropical convection in low vertical wind shears: The role of water vapor. J. Atmos. Sci.,58, 529–545.

  • ——, and G. C. Craig, 1998: Radiative-convective equilibrium in a three-dimensional cloud ensemble models. Quart. J. Roy. Meteor. Soc.,124, 2073–2097.

  • ——, and ——, 1999: Sensitivity of tropical convection to sea surface temperature in the absence of large-scale flow. J. Climate,12, 462–476.

  • Udelhofen, P. M., and D. L. Hartmann, 1995: Influence of tropical cloud systems on the relative-humidity in the upper troposphere. J. Geophys. Res.,100, 7423–7440.

  • Waliser, D. E., 1996: Formation and limiting mechanisms for very high sea surface temperature: Linking the dynamics and the thermodynamics. J. Climate,9, 161–188.

  • ——, and N. E. Graham, 1993: Convective cloud systems and warm-pool sea-surface temperatures—Coupled interactions and self-regulation. J. Geophy. Res.,98, 12 881–12 893.

  • Wallace J. M., 1992: Effect of deep convection on the regulation of tropical sea surface temperature. Nature,357, 230–231.

  • Woolnough, S. J., J. M. Slingo, and B. J. Hoskins, 2000a: The organisation of tropical convection by intraseasonal sea surface temperature anomalies. Quart. J. Roy. Meteor. Soc., in press.

  • ——, ——, and ——, 2000b: The relationship between convection and sea surface temperature on intraseasonal timescales. J. Climate,13, 2086–2104.

  • Yoneyama, K., and T. Fujitani, 1995: The behavior of dry westerly air associated with convection observed during the TOGA-COARE R/V Natsushima cruise. J. Meteor. Soc. Japan,73, 291–304.

  • Zhang, C., 1993: Large-scale variability of atmospheric deep convection in relation to sea surface temperature in the tropics. J. Climate,6, 1898–1913.

  • Zhang, G. J., V. Ramanathan, and M. J. McPhaden, 1995: Convection-evaporation feedback in the equatorial Pacific. J. Climate,8, 3040–3051.

  • View in gallery

    Hovmöller rainfall plot for the CRM experiment. Surface rainfall is summed across the short 64-km axis. The light and dark shading represent a rain rate of 0.1 mm hr and 1 mm hr−1, respectively. For days 0–5 a sine wave SST is imposed with the maximum SST at x = 256 km. This is reversed on day 5

  • View in gallery

    Vertical section taken on day 10 of the normalized water vapor perturbation, q′/σ(q), averaged across the short 64-km axis, where q′ is the water vapor perturbation about the horizontal mean, and σ(q) represents the standard deviation of q. The contours show where the total cloud mass mixing ratio (ice + liquid cloud) averaged across the short axis equals 10−7 kg kg−1

  • View in gallery

    Schematic of interaction between deep convection and SST in the Tropics. The cloud shows the location of convection, the large red arrows show the mean large-scale circulation, and the striped bar shows schematically the SST evolution with red colors representing the warmest SSTs and blue the coolest SSTs. The small arrows near the surface represent surface latent heat fluxes, with the thickness indicating the relative magnitude of the mean fluxes. The total timescale of the SST progression is several weeks

  • View in gallery

    Surface rainfall, water vapor (qυ) at 1520 m, θe at 50 m, and SST at day 10 of the experiment. The solid lines are the average and the dotted lines the maximum value across the 64-km y axis

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 111 111 31
PDF Downloads 124 124 38

On the Relationship between Tropical Convection and Sea Surface Temperature

View More View Less
  • 1 Max-Planck-Institut für Meteorologie, Hamburg, Germany
© Get Permissions
Full access

Abstract

Tropical observations show convective activity increasing sharply above sea surface temperatures (SSTs) of around 26°C and then decreasing as the SST exceeds 30°C, with maximum observed SSTs of around 32°C.Although some aspects of this relationship are reasonably well understood, as of yet no theory has explained the decrease in convective activity above 30°C. Here it is suggested that this aspect of the relationship could result from a organizational positive feedback, sometimes termed “self aggregation,” whereby the occurrence of convection makes future convection more likely to occur in the same location. Using cloud-resolving simulations, it is shown that the feedback between convection and the water vapor field is a good candidate for this role.

Current affiliation: ECMWF, Reading, United Kingdom.

Corresponding author address: Dr. A. M. Tompkins, ECMWF, Shinfield Park, Reading RG2 9AX, United Kingdom.

Email: tompkins@ecmwf.int

Abstract

Tropical observations show convective activity increasing sharply above sea surface temperatures (SSTs) of around 26°C and then decreasing as the SST exceeds 30°C, with maximum observed SSTs of around 32°C.Although some aspects of this relationship are reasonably well understood, as of yet no theory has explained the decrease in convective activity above 30°C. Here it is suggested that this aspect of the relationship could result from a organizational positive feedback, sometimes termed “self aggregation,” whereby the occurrence of convection makes future convection more likely to occur in the same location. Using cloud-resolving simulations, it is shown that the feedback between convection and the water vapor field is a good candidate for this role.

Current affiliation: ECMWF, Reading, United Kingdom.

Corresponding author address: Dr. A. M. Tompkins, ECMWF, Shinfield Park, Reading RG2 9AX, United Kingdom.

Email: tompkins@ecmwf.int

Deep convection in the Tropics is intimately tied to the general dynamical circulation. Thus to understand the observed relationship between convection and sea surface temperature (SST) in the Tropics is also to gain a general understanding of tropical dynamics. The observations show that above a “threshold” SST of 26°C, convective activity increases sharply (Graham and Barnett 1987; Waliser and Graham 1993; Zhang 1993). By sorting the observations into classes of large-scale dynamical activity (i.e., convergence or divergence regimes), Lau et al. (1997) and Bony et al. (1997) revealed that this increase is due to the horizontal SST gradient, and in fact convection is largely independent of the absolute value of the local SST. This has been confirmed using cloud-resolving models (CRMs; Lau et al. 1994;Tompkins and Craig 1999). Thus we can view convective activity and the general tropical circulation as synonymous; an increase in convective activity must be concurrent with an increase in mean atmospheric ascent.

The observations also show that SST is limited to about 32°C, and the fact that temperatures in the extensive “warm pool” region of the western Pacific are within one or two degrees of this maximum has led to the suggestion of a number of SST thermostat mechanisms (i.e., negative feedbacks), involving cloud radiative forcing, surface latent fluxes, and atmospheric and ocean dynamics (Newell 1979; Ramanathan and Collins 1991; Fu et al. 1992; Wallace 1992; Clement et al. 1996;Sun and Liu 1996). All current thermostat theories involve convection. For example, Ramanathan and Collins (1991) proposed that cirrus clouds associated with convection regulate SST by the reduction of the surface incident shortwave (SW) radiation. Wallace (1992) states that surface fluxes and the large-scale flow will control SSTs even in the absence of SW cloud forcing. However, convection is still invoked to “vent” the high SST area, and prevent the boundary layer (BL) and surface temperatures from warming in unison. The role of convection via both enhanced surface fluxes and SW forcing has recently been emphasized (Sud et al. 1999). Other theories involve convection in ocean mixed layer dynamics (Anderson et al. 1996) or as part of a coupled large-scale circulation that forces the upwelling of cool waters (Clement et al. 1996; Sun and Liu 1996).

This leads us to the third aspect of the convection–SST relationship, namely the reduction in convective activity above SSTs of around 30°C. As documented by Waliser (1996), these warmest SSTs often occur within the Pacific warm pool region or the Indian Ocean, with timescales of weeks to a few months, and are termed SST “hot spots” or “warm anomalies.” These regions of warmest SST are often free from convection. Thus the argument is that a lack of local convection (for whatever reason) allows the SST to increase due to increased solar insulation and reduced surface latent heat fluxes. The SST does not run away, however, since it is eventually controlled by the thermostat mechanisms involving convection.

However, as it stands this argument leads to an apparent paradox. If convection is generally increasing with SSTs above 26°C, and SSTs are efficiently limited by convective thermostats, how can developing SST hot spots remain free from convection for periods of weeks to months, allowing the SST anomaly to amplify? It is sometimes claimed (e.g., Waliser 1996; Lau and Sui 1997) that “remotely forced” atmospheric descent provides the mechanism for convective suppression. This leads to a picture where a negative feedback thermostat prevents runaway SSTs, with large-scale flow variability shutting off convection, allowing hot spot development.

But there is a problem with this view. To illustrate this, we use the example of an incipient SST hot spot. For current thermostat theories to work, the BL above this warm patch must be reasonably closely related to the surface thermodynamically, although the water vapor difference between the surface and atmosphere is likely to increase due to low mean winds in the suppressed region (Zhang et al. 1995). Thus under current theories, one would expect positive SST anomalies to be accompanied by higher values of boundary layer equivalent potential energy (θe). The reason why this high θe does not lead to deep convection, removing the SST perturbation on the timescale of a day or less, is that remote forcing prevents this. This seems quite reasonable until we realize that we are deceiving ourselves with a subtle terminology change. What remote forcing actually refers to is “remote convection,” since we have seen that convection and large-scale flow are commensurate. Thus, the assertion that hot spots are the result of remote forcing implicitly contains the assumption that some, as yet undetermined, process ensures that regions over remote cooler SSTs remain more favorable to deep convection over a period of time long enough to allow hot spot development. Either some unspecified process prevents the θe of BL air over the warmest SSTs from exceeding that of remote cooler SST regions, or if this is not the case, the BL θe air over the warmest SSTs is somehow prevented from forming deep convection. Convection over land, for example associated with the Asian monsoon, can not be invoked as an alternative remote forcing candidate, since there is no obvious reason that convection over higher SSTs will be suppressed in preference to lower SST areas.

To shed light on this problem, an idealized experiment is conducted, using a 3D model with a 2-km horizontal resolution to resolve the dynamics of convection, and that represents many ice and warm rain microphysical processes that occur in clouds. The model uses a domain of 1024 km by 64 km in the horizontal, and 20 km deep, with horizontal periodic boundary conditions. The horizontal domain is a compromise enforced by computational limitations. The long 1024-km axis allows reasonably large-scale circulations to be represented in a CRM framework, while the limited third dimension ensures that the organizational scales are not artificially affected by 2D geometry (Tompkins 2000). Details of the model are documented in Tompkins and Craig (1998) and references therein, while the experimental setup is discussed in more detail in Tompkins (2001).

A uniform 2 K day−1 cooling provides the forcing for convection. An underlying SST gradient is imposed along the 1024-km axis, taking the form of one sine wave with SSTs ranging from 299.5 to 300.5 K, giving SST gradients comparable to those observed along the equatorial Pacific. The SST gradient is imposed for a period of 5 days, in order to establish convection over the warmest SSTs. Over the cooler SSTs, convection should become suppressed by mean subsidence. Thus a large-scale overturning circulation is established, modeled for the first time in a 3D framework with resolved deep convection.

Over the coolest SSTs, where deep convection is suppressed, the increase in shortwave incoming flux, and the reduction of surface latent heat flux, would result in a surface flux imbalance. Within the warm pool and Indian Ocean regions, this imbalance can cause SSTs to increase (e.g., Lau and Sui 1997; Woolnough et al. 2000b), perhaps eventually creating a SST warm anomaly. We simulate the warm anomaly development in a very idealized manner, by simply reversing the SST gradient at day 5 of the experiment, to create a surrogate hot spot. Although extremely idealized, this experiment allows us to examine the atmospheric dynamical response to SST anomalies, in an framework where both large- and convective-scale 3D circulations are explicitly represented.

Figure 1 shows the surface precipitation, averaged across the 64-km axis, revealing the location of convection during the experiment. After an initial period of random convection, the large-scale circulation is established during the first 5 days, with ascending motion over the warmest SSTs as expected. After the SST reversal on day 5, the convection dies out quickly over the cool SSTs. However, convection does not spontaneously flare up over the new SST hot spot, but instead propagates slowly toward it. Computing resources limit the experiment to ten days, but judging from the propagation speeds, at least two weeks would be required for the convection to reach the highest SSTs.

Closer examination of the thermodynamic fields reveals a possible reason for this—the interaction with water vapor. Figure 2 shows how, even by day 10, the region centered at x = 768 km, that was free from convection during the first 5 days of the experiment, is very dry. It is well known that convection locally moistens its environment by detraining water vapor and cloud condensate. On the other hand, the subsidence drying is spread out quickly over the deformation radius by gravity waves. Even if an ensemble of cumulus clouds has no net effect on atmospheric water vapor, the regions local to convection are moistened while remote clear-sky areas are dried (see discussion by Randall and Huffman 1980). This is seen directly in observations (Udelhofen and Hartmann 1995; Liao and Rind 1997).

Dry air acts to suppress convection. Firstly, dry air reduces in-cloud buoyancy when entrained into updrafts. Additionally, dry boundary layers (either resulting from convective downdrafts on a local scale, or due to dry air far from convection subsiding into the BL) are associated with low θe values that are less likely to form deep convective events. Observations show examples of intrusions of extratropical dry air suppressing convection (Numaguti et al. 1995; Yoneyama and Fujitani 1995), and lower-tropospheric humidity has been shown to be an important precursor for tropical deep convection (Sherwood 1999). Thus it is only a small step to suggest the existence of a positive organizational feedback, sometimes referred to as “self-aggregation,” where convection locally moistens its atmosphere, making it favorable for future convection.

Adding a positive organizational feedback between convection and water vapor allows us to attempt an explanation for the tropical observations that is illustrated schematically in Fig. 3. When we now invoke atmospheric variability that suppresses convection, the local atmosphere above the BL is also dried (Fig. 3a), since it is a long distance from deep convection. Even if the increasing SST creates a hot spot, convection will not break out due to the inhibition of this dry atmosphere, and SSTs can increase further. Eventually the SST will be arrested by increasing surface fluxes of latent heat, offsetting the solar radiation anomaly, although this “brake” is made less efficient by lower surface winds (Fig. 3b). The latent heat flux will not lead to a local runaway greenhouse effect since the moisture is advected laterally into neighboring convecting regions rather than being vented vertically by local convection. As the BL air is advected, surface fluxes further increase the water vapor content, and a shallow convective layer develops, capped by the dry air above. This is visible in Fig. 2, and is more obvious in Fig. 4, which reveals how the dryness of the free troposphere clearly delineates the “drought” rain-free areas from the deep convecting regions, despite the fact that the BL θe is often comparable.

Thus, a quasi-stable circulation exists, but although convection is suppressed by dryness over the highest SST regions, it will propagate preferentially toward them at the rate at which it can moisten the atmosphere. This is because a state with the convection centered over the warmest SSTs is also one of lowest potential energy (i.e., the difference between mean atmospheric and surface temperature is a minimum). On reaching the SST hot spot, both solar and surface latent fluxes act in unison, via the established thermostat mechanisms, and the SST anomaly will disappear on a much faster timescale than the complementary warming phase with which it was established (Fig. 3c). This difference in warming and cooling timescales is documented in observations (Sud et al. 1999), and it is the reason that the mean convective activity reduces with surface temperature with SST > 30°C, since the atmosphere spends more time in the drought warming phase. The two-week hot spot recovery timescale in the model experiment is slightly shorter than observed, which is to be expected, since the horizontal scale of the SST perturbation is limited to a few hundred kilometers, smaller than the typical observed value that exceeds 1000 km.

The idea that the feedback between water vapor and convection can cause self-aggregation has been previously discussed by Esbensen (1978), Johnson (1978), Nicholls and Lemone (1980), and Randall and Huffman (1980), mostly in the context of shallow cumulus development. Here we are extending this idea to suggest that the self-aggregation feedback between deep convection and water vapor could play a fundamental role in the relationship between convection, large-scale tropical dynamics, and the ocean surface. In observations, Brown and Zhang (1997) and Lucas and Zipser (2000) documented substantial differences in low- to midtropospheric moisture between “rainy” and “drought” periods during the Tropical Ocean and Global Atmosphere Coupled Ocean–Atmosphere Response Experiment (TOGA–COARE) but the causal relationship was difficult to determine. However, strong evidence of a two-way positive feedback relationship has recently been found in 3D CRM simulations (Tompkins 2001), and in a 2D CRM experiment with interactive radiation and a swamp ocean Grabowski (2000) observed that convection did not occur over the highest SSTs, but continuously propagated toward them, while new warm anomalies developed away from the convective region, exactly as suggested here. Moreover, Woolnough et al. (2000a) have shown that the feedback between convection and water vapor appears to be important in regulating the strength and propagation speed of the Madden–Julian oscillation, indicating that the feedback operates on larger spatial and longer temporal scales than can be currently simulated in a 3D CRM framework.

In summary, it appears that surface latent fluxes limit SSTs to 32°C, but that this regime is unstable. Eventually convection must propagate into the high SST areas, and shortwave forcing, possibly in conjunction with ocean dynamical effects, provide a lower SST maximum. The closing remarks of Wallace (1992) should be emphasized. None of the thermostat theories based on convection will prevent maximum tropical SSTs from increasing in a hypothetical future climate. The crucial point that we have attempted to add to the debate is that current theories on their own do not explain the reduction in convective activity with SSTs greater than 30°C. With the addition of an organizational positive feedback, an explanation of the full convection–SST relationship and the existence of convection-free SST hot spots can be attempted. The CRM experiment suggested that the feedback between convection and the water vapor field is a strong candidate for this role, but other positive feedback mechanisms exist that could enhance this. For example, convectively generated cold pools can trigger new convective cells (e.g., Simpson 1980), causing convective clustering. There are indications that radiative feedbacks may also organize convection (Tompkins and Craig 1998; Nilsson and Emanuel 1999; Raymond 2000).

A dry convective atmosphere would not permit temperature anomaly development in the Tropics, since the circulation can quickly respond to remove them on a gravity wave timescale. But due to its slow advective adjustment timescale, water vapor stamps a memory on tropical dynamical circulations, making them robust and slow to respond.

Acknowledgments

Comments by M. Latif, E. Roeckner and J. Slingo are appreciated. The U.K. Met Office provided the cloud model. This work was supported by a European Union Marie Curie Fellowship and the Max Planck Society.

REFERENCES

  • Anderson, S. P., R. A. Weller, and R. B. Lukas, 1996: Surface buoyancy forcing and the mixed layer of the western Pacific warm pool: Observations and one-dimensional model results. J. Climate,9, 3056–3085.

  • Bony, S., K.-M. Lau, and Y. C. Sud, 1997: Sea surface temperature and large-scale circulation influences on tropical greenhouse effect and cloud radiative forcing. J. Climate,10, 2055–2076.

  • Brown, R. G., and C. Zhang, 1997: Variability of midtropospheric moisture and its effect on cloud-top height distribution during TOGA COARE. J. Atmos. Sci.,54, 2760–2774.

  • Clement, A. C., R. Seager, M. A. Cane, and S. E. Zebiak, 1996: An ocean dynamical thermostat. J. Climate,9, 2190–2196.

  • Esbensen, S., 1978: Bulk thermodynamic effects and properties of small tropical cumuli. J. Atmos. Sci.,35, 826–831.

  • Fu, R., A. D. DelGenio, W. B. Rossow, and W. T. Liu, 1992: Cirrus-cloud thermostat for tropical sea-surface temperatures tested using satellite data. Nature,358, 394–397.

  • Grabowski, W. W., 2000: Cloud microphysics and the tropical climate:Cloud-resolving model perspective. J. Climate,13, 2306–2322.

  • Graham, N. E., and T. P. Barnett, 1987: Sea surface temperature, surface wind divergence, and convection over the tropical oceans. Science,238, 657–659.

  • Johnson, R. H., 1978: Cumulus transports in a tropical wave composite for phase-III of GATE. J. Atmos. Sci.,35, 484–494.

  • Lau, K.-M., and C. H. Sui, 1997: Mechanisms of short-term sea surface temperature regulation: Observations during TOGA COARE. J. Climate,10, 465–472.

  • ——, ——, M. D. Chou, and W. K. Tao, 1994: An inquiry into the cirrus-cloud thermostat effect for tropical sea-surface temperature. Geophys. Res. Lett.,21, 1157–1160.

  • ——, H.-T. Wu, and S. Bony, 1997: The role of large-scale atmospheric circulation in the relationship between tropical convection and sea surface temperature. J. Climate,10, 381–392.

  • Liao, X., and D. Rind, 1997: Local upper tropospheric/lower stratospheric clear-sky water vapor and tropospheric deep convection. J. Geophys. Res.,102, 19 543–19 557.

  • Lucas, C., and E. J. Zipser, 2000: Environmental variability during TOGA COARE. J. Atmos. Sci.,57, 2333–2350.

  • Newell, R. E., 1979: Climate and the ocean. Amer. Sci.,67, 405–416.

  • Nicholls, S., and M. A. Lemone, 1980: The fair weather boundary layer in GATE: The relationship of subcloud fluxes and structure to the distribution and enhancement of cumulus clouds. J. Atmos. Sci.,37, 2051–2067.

  • Nilsson, J., and K. A. Emanuel, 1999: Equilibrium atmospheres of a two-column radiative-convective model. Quart. J. Roy. Meteor. Soc.,125, 2239–2264.

  • Numaguti, A., R. Oki, K. Nakamura, K. T. N. Misawa, T. Asai, and Y. M. Kodama, 1995: 4–5-day-period variation and low-level dry air observed in the equatorial Western Pacific during the TOGA-COARE IOP. J. Meteor. Soc. Japan,73, 267–290.

  • Ramanathan, V., and W. Collins, 1991: Thermodynamic regulation of ocean warming by cirrus clouds deduced from observations of the 1987 El Nino. Nature,351, 27–32.

  • Randall, D. A., and G. J. Huffman, 1980: A stochastic model of cumulus clumping. J. Atmos. Sci.,37, 2068–2078.

  • Raymond, D. J., 2000: The Hadley circulation as a radiative-convective instability. J. Atmos. Sci.,57, 1286–1297.

  • Sherwood, S. C., 1999: Convective precursors and predictability in the tropical western Pacific. Mon. Wea. Rev.,127, 2977–2991.

  • Simpson, J., 1980: Downdrafts as linkages in dynamic cumulus seeding effects. J. Appl. Meteor.,19, 477–487.

  • Sud, Y. C., G. K. Walker, and K.-M. Lau, 1999: Mechanisms regulating sea-surface temperatures and deep convection in the tropics. Geophys. Res. Lett.,26, 1019–1022.

  • Sun, D.-Z., and Z. Y. Liu, 1996: Dynamic ocean–atmosphere coupling: A thermostat for the tropics. Science,272, 1148–1150.

  • Tompkins, A. M., 2000: The impact of dimensionality on long-term cloud-resolving model simulations. Mon. Wea. Rev.,128, 1521–1535.

  • ——, 2001: Organization of tropical convection in low vertical wind shears: The role of water vapor. J. Atmos. Sci.,58, 529–545.

  • ——, and G. C. Craig, 1998: Radiative-convective equilibrium in a three-dimensional cloud ensemble models. Quart. J. Roy. Meteor. Soc.,124, 2073–2097.

  • ——, and ——, 1999: Sensitivity of tropical convection to sea surface temperature in the absence of large-scale flow. J. Climate,12, 462–476.

  • Udelhofen, P. M., and D. L. Hartmann, 1995: Influence of tropical cloud systems on the relative-humidity in the upper troposphere. J. Geophys. Res.,100, 7423–7440.

  • Waliser, D. E., 1996: Formation and limiting mechanisms for very high sea surface temperature: Linking the dynamics and the thermodynamics. J. Climate,9, 161–188.

  • ——, and N. E. Graham, 1993: Convective cloud systems and warm-pool sea-surface temperatures—Coupled interactions and self-regulation. J. Geophy. Res.,98, 12 881–12 893.

  • Wallace J. M., 1992: Effect of deep convection on the regulation of tropical sea surface temperature. Nature,357, 230–231.

  • Woolnough, S. J., J. M. Slingo, and B. J. Hoskins, 2000a: The organisation of tropical convection by intraseasonal sea surface temperature anomalies. Quart. J. Roy. Meteor. Soc., in press.

  • ——, ——, and ——, 2000b: The relationship between convection and sea surface temperature on intraseasonal timescales. J. Climate,13, 2086–2104.

  • Yoneyama, K., and T. Fujitani, 1995: The behavior of dry westerly air associated with convection observed during the TOGA-COARE R/V Natsushima cruise. J. Meteor. Soc. Japan,73, 291–304.

  • Zhang, C., 1993: Large-scale variability of atmospheric deep convection in relation to sea surface temperature in the tropics. J. Climate,6, 1898–1913.

  • Zhang, G. J., V. Ramanathan, and M. J. McPhaden, 1995: Convection-evaporation feedback in the equatorial Pacific. J. Climate,8, 3040–3051.

Fig. 1.
Fig. 1.

Hovmöller rainfall plot for the CRM experiment. Surface rainfall is summed across the short 64-km axis. The light and dark shading represent a rain rate of 0.1 mm hr and 1 mm hr−1, respectively. For days 0–5 a sine wave SST is imposed with the maximum SST at x = 256 km. This is reversed on day 5

Citation: Journal of Climate 14, 5; 10.1175/1520-0442(2001)014<0633:OTRBTC>2.0.CO;2

Fig. 2.
Fig. 2.

Vertical section taken on day 10 of the normalized water vapor perturbation, q′/σ(q), averaged across the short 64-km axis, where q′ is the water vapor perturbation about the horizontal mean, and σ(q) represents the standard deviation of q. The contours show where the total cloud mass mixing ratio (ice + liquid cloud) averaged across the short axis equals 10−7 kg kg−1

Citation: Journal of Climate 14, 5; 10.1175/1520-0442(2001)014<0633:OTRBTC>2.0.CO;2

Fig. 3.
Fig. 3.

Schematic of interaction between deep convection and SST in the Tropics. The cloud shows the location of convection, the large red arrows show the mean large-scale circulation, and the striped bar shows schematically the SST evolution with red colors representing the warmest SSTs and blue the coolest SSTs. The small arrows near the surface represent surface latent heat fluxes, with the thickness indicating the relative magnitude of the mean fluxes. The total timescale of the SST progression is several weeks

Citation: Journal of Climate 14, 5; 10.1175/1520-0442(2001)014<0633:OTRBTC>2.0.CO;2

Fig. 4.
Fig. 4.

Surface rainfall, water vapor (qυ) at 1520 m, θe at 50 m, and SST at day 10 of the experiment. The solid lines are the average and the dotted lines the maximum value across the 64-km y axis

Citation: Journal of Climate 14, 5; 10.1175/1520-0442(2001)014<0633:OTRBTC>2.0.CO;2

Save